Person verification via eye movement-driven text reading model

The paper presents a reading-based eye movement biometrics model. The model is able to process passages of text and extract metrics that represent the physiological and behavioral aspects of the eye movements in reading. When tested on a database of eye movements from 103 individuals, the model yielded the Equal Error Rate of 10.2%. The proposed method performed better in the template-aging scenario than comparable eye movement-driven biometrics methods.